Author: blass uri
Date: 02:35:51 04/06/99
Go up one level in this thread
On April 06, 1999 at 03:27:23, Harald Faber wrote: >On April 05, 1999 at 17:11:28, blass uri wrote: > >>I do not agree about fritz. >>Fritz can do horrible positional moves at long time control. >>Fritz is optimized for tournament time control and other programs can earn more >>from over-night analysis of a position. >> >>You can look at the games Fritz5.32-Rebel10C(1 hour per move) >>see rebel site. >> >>Fritz is the only program that lost against Rebel in these games 1.5:.5 > >?? Uri? You really count *2* games?? >I thought you'd know better. I know that the result is not significant but I looked at the games and Fritz played very weak. can you explain the following moves: 3rr1k1/p3bpp1/nnp4p/1b1p3P/1p1P1B2/1P1N1PPB/PN1KP3/2R4R b - - 0 1 Fritz5.32 played 23...Ra8 and after 24.e3 Nb8. Can your program find the moves 23...Ra8 and 24...Nb8 if you give it a long time? How much time Did Dritz need to find the moves 23....Ra8? and 24....Nb8? It may need to know the history of the game to find these moves so I give the begining of the game. 1.d4 Nf6 2.c4 e6 3.Nf3 b6 4.g3 Ba6 5.Qa4 c6 6.Nc3 d5 7.cxd5 exd5 8.Bg5 Be7 9.Ne5 b5 10.Qc2 h6 11.Bf4 Qa5 12.Bg2 b4 13.Qa4 Qxa4 14.Nxa4 Nfd7 15.Nd3 0-0 16.Kd2 Re8 17.b3 Bb5 18.Rac1 Na6 19.h4 Rad8 20.h5 Nf6 21.f3 Nd7 22.Bh3 Nb6 23.Nab2 Ra8?? 24.e3 Nb8?? I read that fritz5.32 cannot find these moves at tournament time control so I suspect that time is counter productive for fritz if it significantly slower than 40 moves/2 hours . Fritz5.32 used 128MB hash tables to find the stupid moves. Uri
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.